Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/jerrylalala/compound-engineering/executor-capability-gatenpx skills add Jerrylalala/compound-engineering --skill executor-capability-gategit clone --depth 1 https://github.com/Jerrylalala/compound-engineeringWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00066 | $0.01373 |
| Opus 5 | $0.00033 | $0.00687 |
| Sonnet 5 | $0.00013 | $0.00275 |
| Haiku 4.5 | $0.00007 | $0.00137 |
Grade B, and why
executor-capability-gate scanned grade B with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
LAST_CALL=$(cat ~/.codex/.last_call 2>/dev/null || echo "0") Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -s --max-time 5 "https://api.openai.com" -o /dev/null -w "%{http_code}" How it starts
The opening of the file, as written. The whole thing — 161 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Executor Capability Gate — 外部调用前置检查
Codex 洞察(P8,新增项):调用外部模型前做前置检查比"自动路由"更实用。 5 项检查防止无效调用,是 P7 Codex-first Executor 的前置依赖。
五项前置检查
Check 1: CLI 安装检查
# Codex
command -v codex &>/dev/null
echo "exit: $?" # 0=已安装, 1=未安装
# Gemini
command -v gemini &>/dev/null
失败处理:
❌ Codex CLI 未安装
安装命令:npm install -g @openai/codex
或:bun install -g @openai/codex
Check 2: 登录状态检查
# Codex - 检查凭据文件是否存在(codex --version 无需登录,不能用于验证)
[ -f ~/.codex/auth.json ] && echo "OK" || echo "NOT_LOGGED_IN"
# Gemini
gemini --version 2>&1 | grep -q "version" && echo "OK" || echo "NOT_LOGGED_IN"
失败处理:
❌ Codex 未登录(~/.codex/auth.json 不存在)
登录命令:codex (首次运行引导登录)
Check 3: 网络连通性检查
# 检查网络可达性(仅连通性,不含认证——凭据走 auth.json,非 OPENAI_API_KEY)
curl -s --max-time 5 "https://api.openai.com" -o /dev/null -w "%{http_code}"
# 非 000 = 网络可达(包括 401 均表示网络通)
# 000 = 网络不可达
失败处理:
❌ 网络不可达(curl 返回 000)
跳过 Codex 调用,退回 Claude 执行
Check 4: Rate Limit 检查
# 检查最近 Codex 调用记录(简单本地记录)
# 注意:~/.codex/.last_call 由本 gate 在调用通过后写入(见门控输出格式末尾)
LAST_CALL=$(cat ~/.codex/.last_call 2>/dev/null || echo "0")
NOW=$(date +%s)
ELAPSED=$((NOW - LAST_CALL))
if [ $ELAPSED -lt 60 ]; then
echo "RATE_LIMITED: 距上次调用 ${ELAPSED}s,建议等待至少 60s"
fi
# 调用通过后,写入时间戳(防止频繁调用):
# echo $(date +%s) > ~/.codex/.last_call
重要:调用 Codex 成功完成后,必须执行 echo $(date +%s) > ~/.codex/.last_call 以更新记录,否则 rate limit 检查永远通过(文件不存在时 ELAPSED 极大)。
Check 5: 任务适配性检查
根据任务特征快速判断(详细决策逻辑见 codex-first-executor skill):
| 任务特征 | Codex 适合? |
|---|---|
| 大量机械 patch(格式化、重命名) | ✅ 适合 |
| 高风险改动(auth、payment、migration) | ❌ 不适合 |
| 纯分析/research 任务 | ✅ 适合 |
| 视觉/UI 任务 | ❌ 不适合 |
| 需要项目上下文的重构 | ⚠️ 谨慎 |
门控输出格式
每次外部调用前输出检查结果:
🔍 Executor Capability Gate — Codex 检查
✅ CLI 已安装 (codex v0.1.x)
✅ 已登录
✅ 网络正常 (API 200)
✅ Rate limit 正常 (距上次 120s)
✅ 任务适合 Codex(批量 patch)
→ 允许调用 Codex
或:
🔍 Executor Capability Gate — Codex 检查
✅ CLI 已安装
❌ Rate limit(距上次仅 30s)
⚠️ 任务高风险(涉及 auth/payment)
→ 跳过 Codex,由 Claude 执行
理由:rate limit + 高风险任务不适合外部执行器
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 161 lines · 66 tokens per session scan B 67fcf84a7700
executor-capability-gate is a skill published in the GitHub repository Jerrylalala/compound-engineering (5 stars, last pushed 3mo ago), licensed MIT. It adds 66 tokens to every session and 1,373 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it B with 2 findings (reads agent configuration directories, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…